Author:
He Wenyang,Zhao Wenlong,Jiang Yuan
Abstract
Abstract
Machine game is one of the important directions of artificial intelligence research, and Chinese chess is a typical game process. This paper first introduces the game principle of Chinese chess. Then Q-learning method and evaluation function are added to train data through a large amount of self-learning. A chess game system based on Q-learning is designed and developed. The experimental results show that Chinese chess with Q-learning algorithm has the ability of evolutionary learning, which effectively improves the game level of Chinese chess.
Reference12 articles.
1. Learning to Predict by the Methods of Temporal Differences [J];Sutton;Machine Learning,1988
2. Using a Computer Game to Develop Advanced AI;Laird;Computer,2001
3. Machine Learning - Artificial Intelligence; Study Results from University of Saarland in the Area of Artificial Intelligence Reported (Dynamics in matching and coalition formation games with structural constraints) [J],2018
4. Human-Level Control through Deep Reinforcement Learning;Mnih;Nature,2015
5. Q-learning [J];Dayan;Machine Learning,1989
Cited by
3 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献